FORECASTING DEVICE, PARAMETER SET PRODUCTION METHOD AND PROGRAM
|Posted date||Apr 19, 2018|
|International application number||2017JP025236|
|International publication number||WO 2018012487|
|Date of international filing||Jul 11, 2017|
|Date of international publication||Jan 18, 2018|
|Title||FORECASTING DEVICE, PARAMETER SET PRODUCTION METHOD AND PROGRAM|
|Abstract||The purpose of the present invention is to provide a forecasting device capable of also achieving long-term forecasting with high accuracy by using a large-scale time-series data stream. In the case of the forecasting device 1, a count window calculation unit 11, a regime update unit 13, and a regime addition unit 15 correspond to each hierarchy of a hierarchical structure of a data stream. Furthermore, a numerical model specified by a parameter set stored in a parameter set storage unit 7 includes a non-linear component, and is thus able to express non-linearity of the data stream. The regime update unit 13 updates the parameter set, thereby achieving forecasting by means of a non-linear dynamic system. Furthermore, the regime addition unit 15 adds a new pattern (regime) to the data stream. The regime update unit 13 utilizes regime shift, that is, transition from a certain regime to another regime, in an event stream. Consequently, highly accurate long-term forecasting can also be achieved.|
|Scope of claims||
1. At time tcto part of the time-series data in the current window X XCt time usingcl fromsafter the step of predicting the value of the one or more of the event in the device, a set of parameters of the storage means, and means for updating the regime, prediction means, said storage means is a set of parameters, identifying the stored set of parameters of the mathematical model, the mathematical model, a non-linear element, wherein the set of parameters, wherein the non-linear element comprises a non-linear parameters to determine coefficients of a, wherein the updating means is regime, without changing said non-linear parameters, the set of parameters of the part or all of the other parameters included in the update, in the current window XCat each time of the data, the updated set of parameters of the mathematical model specified by the current window obtained using XcV event value corresponding to each time ofCto reduce the difference, the prediction means includes, the updated set of parameters of the mathematical model specified by the time t usingcl fromsafter the step of predicting the value of the one or more of the event, the prediction device.
2. Wherein the storage unit stores a set of parameters, c (c is a natural number) a set of parameters of two θi(i=1,...,c) is stored, wherein the predicting means, the post-update c of a set of parameters of θisome or all of the value of the event using VEprediction, the prediction device according to claim 1.
3. Means for adding regime, said regime additional means, in the current window XCat each time point of the data, the updated c of a set of parameters of θiis obtained using at each time corresponding to the event value VCis different from the additional conditions are met, the set of parameters of the storage means, a new parameter set θc+1is added, said regime update means updates, c+1 are a set of parameters of θi(i=1,...,c+1) for non-linear parameters other than parameter part or all of the update, wherein the predicting means, a set of parameters of the updated c+1 θi(i=1,...,c+1) some or all of the mathematical model is specified by using the estimation of a value of an event, the prediction device according to claim 2.
4. Wherein the mathematical model, a linear elements, wherein the set of parameters, wherein the identifying comprises of the linear parameters of the linear elements, wherein the regime additional means, without changing said non-linear parameters and the linear parameter is determined, the above-mentioned linear determined using the parameters to determine the parameters of the non-linear, from any of the claim 1 3 with the prediction device.
5. X in the current windowc(j) (j=1,...,h、his a natural number) is, h hierarchy and, wherein the parameter sets is, wherein the level of the hierarchy in the current window X hc(j) h and corresponds to the hierarchy, wherein the regime and the updating unit, updating said set of parameters of each layer, wherein the predicting means, each of the levels from the value predicted in the event the event to predict the value of the entire, any one of claims 4 claim 1 from with the prediction device.
6. At time tcup to the time-series is part of the data in the current window XCis used, based on a mathematical model parameter for specifying a part of the set parameter is changed to a new parameter set to produce a set of parameters of the production methods, the mathematical model, the non-linear element and comprises, as the parameter sets, wherein the non-linear elements specifying the non-linear control parameters, the information processing device in accordance with the regime update means, wherein said non-linear parameters do not change, the parameters included in the set and other parameters of all or part of the updated, in the current window XCat each time of the data, the updated set of parameters identified by the mathematical model is obtained by using in the current window XCV event value corresponding to each time ofCand update to reduce a difference in a set of parameters comprises the step of a method for the production.
7. The computer, claim 1 from any one of claims 5 to function as a prediction of a program.
|IPC(International Patent Classification)|
National States: AE AG AL AM AO AT AU AZ BA BB BG BH BN BR BW BY BZ CA CH CL CN CO CR CU CZ DE DJ DK DM DO DZ EC EE EG ES FI GB GD GE GH GM GT HN HR HU ID IL IN IR IS JO JP KE KG KH KN KP KR KW KZ LA LC LK LR LS LU LY MA MD ME MG MK MN MW MX MY MZ NA NG NI NO NZ OM PA PE PG PH PL PT QA RO RS RU RW SA SC SD SE SG SK SL SM ST SV SY TH TJ TM TN TR TT TZ UA UG US UZ VC VN ZA ZM ZW
ARIPO: BW GH GM KE LR LS MW MZ NA RW SD SL SZ TZ UG ZM ZW
EAPO: AM AZ BY KG KZ RU TJ TM
EPO: AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR
OAPI: BF BJ CF CG CI CM GA GN GQ GW KM ML MR NE SN ST TD TG
Contact Information for " FORECASTING DEVICE, PARAMETER SET PRODUCTION METHOD AND PROGRAM "
- National University Corporation Kumamoto University Social Relations Section Research Promotion and International
- URL: https://www.kumamoto-u.ac.jp/kenkyuu_sangakurenkei
- Address: 2-39-1, Kurokami, Kumamoto-shi, Kumamoto-ken, Japan , 860-8555
- Phone: 81-96-342-3277
- Fax: 81-96-342-3239